SS early or delayed- working with numbers

[kathyc2]

We've had a lot of discussions here about if it's better to take SS early or wait. I put a speadsheet together for working it out, and I'm quite surprised how the numbers are coming out. Would someone who understands finance and spreadsheets double check me?

Here's the assumptions I started with:
Used COLA of 2%
FRA is 66, at which time benefits would be 25,000.
At 64 benefits would be 21,500 (25000 x .86). Using the 2% COLA by age 90 I show total benefits to be 759,903.
Waiting until age 70 benefits would be 33,000 (25000 x 1.32). I then also applied the 2% COLA for the 6 years so that at age 70 benefits would be 37,163. Continuing using the 2% COLA by age 90 cumulative benefits would be 958,195, or 198,292 higher benefits than starting at age 64.
That is pretty much the number I think we would all expect.

Then I calculated the difference of if TP waits until age 70, but take out of retirement savings the amount he would have received in SS starting at age 64. For example, at age 64 he takes 21,500 from retirement account, at age 65 he takes out 21,930 (21500 x 1.02), and continues this for the 6 years until he starts taking benefits at age 70. For the retirement account I used a rate of return of 6%. I'm coming up with by age 90 he will have 534,213 more in retirement account if he starts SS at 64 and does not take the corresponding amount from retirement accounts. I have to take the rate of return down to 1.65% to get it to be the same as the "extra" SS benefits received by waiting to age 70 instead of 64.

I didn't expect the difference to be this large, so would appreciate if someone checks my math....

I realize there are other factors, such as if TP continues to work or not past 64, taxes, etc. Can we not bring all that into account right now and just focus on if the numbers are correct?

Thanks.

[FEDUKE404]

Comparing benefits and making "best" choice for Social Security & retirement income

I will let someone else crunch the numbers. I'm sure you're quite adept at that aspect.

Your final statement is everything in a nutshell. . .the "other things."

Determining rates of return and COLAs, especially that far out, can be treacherous. You have already (kinda) mentioned "taxes." That aspect alone would seem to be the largest variable. Things that come to mind include how much of Soc Sec benefits would be subject to tax (85% can always be changed), tax rates/deductions, additional Obamacare taxes on investment income, where is AMT, and my favorite complaint IRMAA.

And, for most people, the truly unknown factor of one's future health or the lack thereof. Spreadsheets are of little assistance there.

It's been my observation that a large number of people take Soc Sec at the first opportunity (age 62). Many of them never worked through "the numbers." I know of others adamantly waiting until age 70. More power to them! I went at full-retirement age of 66 and figured that to be a reasonable compromise. And to keep IRS / IRMAA somewhat at bay, I'm already taking some 401k distributions although RMDs are not here quite yet.

My 2¢ worth: You can prognosticate all you want about pure numbers, but there are simply too many variables that also come into the mix. It is admirable to run those numbers, but at some point it is always possible all of your (your client's) plans can go up like a puff of smoke.

I'll continue to read this thread and look forward to what others may say.

FE

[Rapid Robert]

Here's the assumptions I started with:

Using the 2% COLA by age 90

I'm coming up with by age 90

Note: I will try to deal with kathyc2's numbers in a subsequent reply.

As an observation, not a direct response to the question, (and not a criticism!), I don't understand why it seems people are always using such unrealistic numbers for life expectancy. It's my understanding that the SS benefits formula is supposed to be actuarially neutral. So why does everyone always only look at people who exceed the average? Why don't we also see the parallel calculation for someone who only lives to age 78? Only about one fourth of 65 year olds alive today will make it to 90, so why pay much attention to that?

Better yet, why not just use the actual life expectancy number?

https://www.ssa.gov/planners/lifeexpectancy.html

It's just plain ridiculous to say people are better off waiting as long as possible to start, yet 99% of the articles and columns you see out there always seem to say that.

Maybe it's like we're living in Garrison Keillor's fictional town Lake Wobegon, "where all the women are strong, all the men are good looking, and all the children are above average".

As a side note, as a male who is about 2.5 years younger than my wife, I really get tired of all the assumptions and implications I read that somehow it's way more important to plan for her survivor benefits than mine. I'm just as likely to survive her as the other way around.

[Rapid Robert]

Then I calculated the difference of if TP waits until age 70, but take out of retirement savings the amount he would have received in SS starting at age 64. For example, at age 64 he takes 21,500 from retirement account, at age 65 he takes out 21,930 (21500 x 1.02), and continues this for the 6 years until he starts taking benefits at age 70. For the retirement account I used a rate of return of 6%. I'm coming up with by age 90 he will have 534,213 more in retirement account if he starts SS at 64 and does not take the corresponding amount from retirement accounts.

First let me try to re-phrase your scenario, since I kind of got lost with so many scenarios.

Person needs annual income starting at age 64. He can either start taking SS benefits right away, or he can take equivalent money out of pre-tax retirement investments for six years and then replace that income with SS benefits beginning age 70.

It's no surprise that his retirement account will be much larger if he does not spend six years draining a large chunk of it away. What I think is missing, is that under the case where he starts taking SS at age 64, you should begin at age 70 taking enough money out of retirement account every year to makeup the difference between his lower SS benefit, and what it would have been if he waited to age 70. Wouldn't that bring the retirement account balance down more in line with what you expected?

[kathyc2]

What I think is missing, is that under the case where he starts taking SS at age 64, you should begin at age 70 taking enough money out of retirement account every year to makeup the difference between his lower SS benefit, and what it would have been if he waited to age 70. Wouldn't that bring the retirement account balance down more in line with what you expected?

Ah! I knew I had to be doing something wrong…..

So now I have 2 IRA projections:
“Delayed” take the SS that could be taken for 6 years out of IRA.
“Early” does not take any money out of IRA for first 6 years, and then from year 7 on takes out the difference of projected higher benefits that could have been received if waited less the amounts received by starting at 64.

Using 6% rate of return, early is higher until age 86, at which time delayed becomes higher.

It should be noted that using different COLA and rate of return numbers would give different results, as would for people who’s FRA is greater than 66.

[JohnH]

There are many calculators out there for running the numbers, and many points of view on the issue. Some are driven by personal opinion and others are driven by what the person doing the analysis is trying to sell (Financial Advisors selling annuities & brokerage services quickly come to mind). The results depend greatly upon the up-front assumptions and can vary widely.

It is clear that waiting provides the greatest benefit, PROVIDED that the person lives until the final day of their life expectancy or greater. The actuaries design their tables that way - that's their job. Its hard to argue with a guaranteed 6-8% return on your investment, which is essentially what you get by delaying. But if one delays and then has the bad judgement to die younger than their projected life expectancy, it was in retrospect a bad decision. However, I don't think people will be standing around the casket saying "Looks like John was a chump - he left some Social Security benefits lying on the table."

On the other hand, how many of us prepare tax returns for elderly people who just don't have enough income to live comfortably? Many of them had a decent income at the time they retired, but inflation and other unknowns eroded the value of that fixed income. To me, that is the greatest risk - outliving one's income sources and spending their final years living in poverty. It is MUCH more of a risk than dying early and leaving a few thousand of benefits unclaimed.

Some people have no choice. They lose their jobs, their health fails, etc, and so they must begin drawing Social Security benefits at the earliest possible age. Others make what I consider to be a silly decision based on the reasoning that "I've paid into it all these years - its time to get some back." One could just as easily say it's useless to pay my insurance premiums because I've never filed a claim. There's no logic behind either statement - it's simply emotion and opinion based on faulty reasoning.

For anyone who is able to work and doesn't have any known life-shortening illnesses, I think it is a monumental mistake to begin drawing Social Security benefits prior to full retirement age (around 66), unless they have considerable financial resources backing up that decision. The decision to delay past full retirement age (delaying to age 70 or some point in between), is a bit more murky, but it's the norm for high-net worth people to do so if they make a purely financial decision. So to me, the key is to make a financial decision and leave the emotion out of it.

Like FEDuke, I waited until FRA, mainly because I was still working full time and the payback of benefits coupled with the tax bite made it unwise. I did decide not to delay until age 70, but the wisdom of that decision will not be known unless I live to age 88-90.
And even if it turns out to have been a bad decision, I doubt I'll have to live with the consequences very many years past THAT break-even point.

[kathyc2]

I think we have all run into situations where clients that want to retire early are ill prepared.

I’m working on an actual situation where that is not the case. They both retired earlier this year, ages 63 and 64. They have no debt, LTC insurance and about 1M set aside; 75% traditional and 25% Roth. Something very extreme would need to occur for them to not be able to live the lifestyle they desire.

Her SS benefits are quite a bit lower and will start Jan 18. Flexible on when he will start.

Meeting next week to get actual numbers for 2017 to see how much they can convert to Roth with the goal of having taxable income at the cusp of where 15% marginal ends. Since his RMD’s will start 2 calendar years earlier, will convert all from his.

One option is to have him wait a year or 2 to draw benefits and draw down/convert traditionals to where RMD’s do not exceed inflation adjusted needed amount and also to keep less SS taxable and keep in the 15% projected marginal rates.

Lucky for me, they are both pretty good at understanding numbers and spreadsheets, so as long as I can plot out different projections, they are quite capable of making an informed decision themselves. I and they realize that projections are only projections and anything can change.

[Rapid Robert]

On the other hand, how many of us prepare tax returns for elderly people who just don't have enough income to live comfortably? Many of them had a decent income at the time they retired, but inflation and other unknowns eroded the value of that fixed income. To me, that is the greatest risk - outliving one's income sources and spending their final years living in poverty.

I don't know -- how many? I don't have any clients like that. If they don't have much income, why are they filing a tax return?

What income did they have that was "fixed" at retirement? Soc. Sec. is not fixed. Taking money out of pre-tax accounts is not fixed. The equity in their house if they have one is not fixed. Maybe some defined benefit pension plans are fixed, but those are available to fewer and fewer retirees all the time.

It seems like what we read about most often is people who are deliberately trying to be impoverished so they can get on Medicaid. Aren't they often trying to figure out how to have cake and eat it too, by keeping the house in the family but not counting it as an asset?

What about adult children? Not all, but certainly quite a few seniors ought to be able to rely at least a little on support from family members if they are really that hard up. That's the way Americans used to do it -- I had a grandparent move in when I was five years old, and more recently my father in law moved in for his last few years while we still had young kids living at ahome -- and neither my family of origin nor my current family is what you'd call rich. And if the elderly never had kids, well then they sure should have been able to save up a ton of money!

[mactoolsix]

Matching RMD to Expenses

Kathy - the goal should be to get the traditional RMDs down close to their projected living expenses. Converting traditional to a Roth and then having to use Roth funds in a few years means they may be paying income tax prematurely.

On the other hand we have no idea how Trump's new tax proposals will effect them in the future. I have struggled with this personally and with a few clients over the years. Just when I think I have it figured out, the tax law changes, or worse yet the client decides to sell the family home with tons of capital gain and downsize!

It's like a bouncing football ball . . . .

Mike

[kathyc2]

Kathy - the goal should be to get the traditional RMDs down close to their projected living expenses. Converting traditional to a Roth and then having to use Roth funds in a few years means they may be paying income tax prematurely.


Mike

I'm not so sure I see it that way. I'm more concerned with the effective rate rather than when it's paid. The higher the RMD, the less variables to do tax planning with down the road. For instance if the RMD is 50K down the road, there is no choice but to take that and pay whatever related tax it produces. Now, if instead, the RMD is 25K, they can choose to take more than that if it produces a better tax alternative. Say 40K produces the optimized tax situation, but they only need 30K of that to meet current expenses. They can then convert the extra 10K and eliminate tax on it rather than deferring.

Like you said, we don't know what tax changes may be coming in the next 30 years or so. That's why having more options than a high RMD are important, IMO.

[Rapid Robert]

Converting traditional to a Roth and then having to use Roth funds in a few years means they may be paying income tax prematurely.

That's kind of funny, when you think of it. The Trad. IRA tax has already been deferred for many years if not decades, yet you describe finally paying the tax as being "premature".

[FEDUKE404]

Future numbers, again

I'm not so sure I see it that way. I'm more concerned with the effective rate rather than when it's paid. The higher the RMD, the less variables to do tax planning with down the road. For instance if the RMD is 50K down the road, there is no choice but to take that and pay whatever related tax it produces. Now, if instead, the RMD is 25K, they can choose to take more than that if it produces a better tax alternative. Say 40K produces the optimized tax situation, but they only need 30K of that to meet current expenses. They can then convert the extra 10K and eliminate tax on it rather than deferring.

Like you said, we don't know what tax changes may be coming in the next 30 years or so. That's why having more options than a high RMD are important, IMO.

Is there a possibility your (apparent) assumption that the future RMDs will be constant may be in error?

A decrease in account assets due to withdrawals could change the RMDs, and of course if there is a stock market decline there would likely be a corresponding reduction in RMD amounts based upon 12/31/20XX values. Of course, a charging bull market could have the opposite effect.

I think there is a serious difference between running an analysis of fixed numbers versus a similar analysis with a collection of "unknowns" becoming an integral part of the calculations.

It's kinda like I told a client today: "If you can let me know what day you plan to die and how the stock market will perform between now and then, I can give you a pretty fair analysis of a 'wise' tax strategy."

FE

[kathyc2]

Is there a possibility your (apparent) assumption that the future RMDs will be constant may be in error?

A decrease in account assets due to withdrawals could change the RMDs, and of course if there is a stock market decline there would likely be a corresponding reduction in RMD amounts based upon 12/31/20XX values. Of course, a charging bull market could have the opposite effect.

I think there is a serious difference between running an analysis of fixed numbers versus a similar analysis with a collection of "unknowns" becoming an integral part of the calculations.

It's kinda like I told a client today: "If you can let me know what day you plan to die and how the stock market will perform between now and then, I can give you a pretty fair analysis of a 'wise' tax strategy."

FE

Of course anytime you project anything out 20-30 years you need to make assumptions based on most likely projections. Of course the numbers will change. As time goes on, the plan needs to change as more things become known rather than projected. I'm not big on "winging it", and someone would have a very difficult time to convince me that it's better to have less options available for future planning rather than more options.

[kathyc2]

IMO the mantra that it's always better to defer has wrongly been ingrained on the public too much and too long. I have several older clients that take no more than their RMD's and as a result after standard/exemptions they in effect have "negative taxable income". Many refuse to see that they could either take more out or even better convert to Roth with zero federal tax. So, instead of taking the "freebee" they leave it for heirs to pay 15-25% or more tax on it.

[kathyc2]

When I was going over options w/ clients today, I was talking about the qualified dividends in their non-qualified plan. Maybe I'm punch drunk w/ too many numbers, but just saying it was funny. No wonder people think accountants speak a different language!